⌦abliteratedmodels.org

Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal

Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal is a abliterated and uncensored variant of AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16, published on Hugging Face by AEON-7. It is 27B parameters, 48GB VRAM class class and other licence. It is distributed across 4 repositories in NVFP4 and MLX formats, totalling 10.2K downloads.

Abliteration + uncensored fine-tune27BQwen25B – 50B↓ 10,224♥ 38
View on Hugging Face ↗Base model ↗

Specification

Model name
Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal
Source repository
AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-MLX-FP4
Base model
AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16
Publisher
AEON-7
Parameters
27B
VRAM required
~17.2 GB at Q4_K_M, 4k context (estimated) — runs on 24 GB cards
Hardware class
25B – 50B — 48GB VRAM class
Licence
other
Task
Image text to text
Context window
256K tokens (as documented)
Ablation scope
Not stated by the publisher
Formats available
NVFP4, MLX
First indexed
28 Apr 2026
Last updated
6 Oct 2026

Ablation technique

How Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal was modified, and what that implies.

Abliteration + uncensored fine-tune

Directional ablation combined with additional fine-tuning on unfiltered data, so behaviour diverges from the base model beyond refusal removal alone.

Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.

From the publisher’s model card

“This is the smallest member of the MLX quant grid (16 GB) and the fastest single-stream — a high-quality compact build for Apple Silicon that fits a 24 GB Mac and, with its native MTP drafter, decodes at 26.5 tok/s — roughly 3.2× the 8-bit's 8.2 tok/s. It stays fully coherent, keeps the abliteration intact, and...”

Running it

Commands are templates — confirm the exact repository and quant file before use.

Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-MLX-FP4"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
    repo, torch_dtype="auto", device_map="auto"
)
vLLM
vllm serve AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-MLX-FP4 --trust-remote-code

Downloads and variants (3)

Every published repository of Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal, including quantised re-releases by other authors.

RepositoryFormatDownloads
AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-MLX-FP4sourceMLX3.5K
AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-NVFP4-MTPNVFP43.4K
AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-MLX-8bitMLX3.3K

Security-team assessment

Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.

  • Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
  • Capability is inherited from AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16, not added by ablation — at 27B parameters, expect roughly the base model’s competence, minus some instruction-following fidelity.
  • The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
  • This release adds fine-tuning on unfiltered data, so its behaviour diverges from the base model beyond refusal removal — differences you measure cannot be attributed to ablation alone.
  • Declared licence is other, but the terms that bind you are AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16’s — a re-release cannot grant more than its parent.

Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.

Metadata for Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with AEON-7, and does not independently verify publisher claims. See responsible use.